diff --git a/README.MD b/README.MD index 6eb0c0f..05cab79 100644 --- a/README.MD +++ b/README.MD @@ -146,7 +146,7 @@ When this error has occurred, please check the network environment. **If the dependency package error after updating, please double clicking ```repair_dependency.bat``` (for Official ComfyUI Protable) or ```repair_dependency_aki.bat``` (for ComfyUI-aki-v1.x) in the plugin folder to reinstall the dependency packages.
- +* Commit [GetMainColorsV2](#GetMainColorsV2) node, add sorting by color area and output color values and proportions in the preview image. This part of the code was improved by @ HL, thanks. * Optimize dependency packages. Optimize some algorithms. * Split some nodes of the dependencies that are prone to problems into [ComfyUI_LayerStyle_Advance](#https://github.com/chflame163/ComfyUI_LayerStyle_Advance) repository. Including: LayerMask: BiRefNetUltra, LayerMask: BiRefNetUltraV2, LayerMask: LoadBiRefNetModel, LayerMask: LoadBiRefNetModelV2, @@ -1099,6 +1099,12 @@ Outputs: * preview_image: 5 main color preview images. * color_1~color_5: Color value output. Output an RGB string in HEX format. +### GetMainColorsV2 +Add sorting by color area to the [GetMainColors](#GetMainColors) node and display color values and color areas in the preview image. +This part of the code was improved by @ HL, thanks. +![image](image/get_main_color_v2_example.jpg) + + ### ColorName Output the most similar color name in the color palette based on the color value. diff --git a/README_CN.MD b/README_CN.MD index 459326f..0d1a52e 100644 --- a/README_CN.MD +++ b/README_CN.MD @@ -127,6 +127,7 @@ os.environ['HF_ENDPOINT'] = 'https://hf-mirror.com' ## 更新说明 **如果本插件更新后出现依赖包错误，请双击运行插件目录下的```install_requirements.bat```(官方便携包)，或 ```install_requirements_aki.bat```(秋叶整合包) 重新安装依赖包。 +* 添加 [GetMainColorsV2](#GetMainColorsV2) 节点，增加按颜色面积排序，并在预览图中输出色值和比例。这部分代码由@HL完善，感谢。 * 优化依赖包。优化部分图形算法。 * 分拆部分依赖易出问题的节点至[ComfyUI_LayerStyle_Advance](#https://github.com/chflame163/ComfyUI_LayerStyle_Advance)仓库。包括下列节点： LayerMask: BiRefNetUltra, LayerMask: BiRefNetUltraV2, LayerMask: LoadBiRefNetModel, LayerMask: LoadBiRefNetModelV2, @@ -973,6 +974,12 @@ ImageScaleByAspectRatio的V2升级版 * preview_image: 5个主色预览图片。 * color_1~color_5: 色值输出。输出格式为HEX格式的RGB字符串。 +### GetMainColorsV2 +在[GetMainColors](#GetMainColors)节点基础上增加按颜色面积排序，并在预览图片中显示色值和颜色面积。 +这部分代码由@HL完善，感谢。 +![image](image/get_main_color_v2_example.jpg) + + ### ColorName 根据色值输出调色盘里最近似的颜色名称。 ![image](image/color_name_example.jpg) diff --git a/image/get_main_color_v2_example.jpg b/image/get_main_color_v2_example.jpg new file mode 100644 index 0000000..6a9d73a Binary files /dev/null and b/image/get_main_color_v2_example.jpg differ diff --git a/py/get_main_colors.py b/py/get_main_colors.py index 945c146..783a256 100644 --- a/py/get_main_colors.py +++ b/py/get_main_colors.py @@ -1,10 +1,152 @@ import torch -from PIL import Image, ImageDraw +from PIL import Image, ImageDraw, ImageFont +from collections import Counter +import colorsys from .imagefunc import AnyType, log, tensor2pil, pil2tensor, load_custom_size, gaussian_blur from .imagefunc import RGB_to_Hex any = AnyType("*") + +class LS_GetMainColorsV2: + + def __init__(self): + self.NODE_NAME = 'Get Main Colors V2' + + @classmethod + def INPUT_TYPES(self): + size_list = ['custom'] + size_list.extend(load_custom_size()) + k_means_algorithm_list = ["lloyd", "elkan"] + return { + "required": { + "image": ("IMAGE",), + "k_means_algorithm": (k_means_algorithm_list,), + }, + "optional": { + } + } + + RETURN_TYPES = ("IMAGE", "STRING","STRING", "STRING", "STRING", "STRING",) + RETURN_NAMES = ("preview_image", "color_1", "color_2", "color_3", "color_4", "color_5",) + FUNCTION = 'get_main_colors_v2' + CATEGORY = '😺dzNodes/LayerUtility' + + def get_main_colors_v2(self, image, k_means_algorithm): + ret_images = [] + grid_width = 512 + grid_height = 64 # Reduced height to fit 10 colors + + for i in range(len(image)): + pil_img = tensor2pil(torch.unsqueeze(image[i], 0)).convert("RGB") + blured_image = gaussian_blur(pil_img, (pil_img.width + pil_img.height) // 400) + + accuracy = 60 + num_colors = 5 # Increased to 5 colors + num_iterations = int(512 * (accuracy / 100)) + original_colors, color_percentages = self.interrogate_colors( + pil2tensor(blured_image), num_colors=num_colors, algorithm=k_means_algorithm, mix_iter=num_iterations, + random_state=0) + + main_colors = self.ndarrays_to_colorhex(original_colors) + + # Sort colors by percentage + sorted_colors = sorted(zip(main_colors, color_percentages), key=lambda x: x[1], reverse=True) + print(f"sorted_colors={sorted_colors},type={type(sorted_colors)}") + # Create color info string with HSB values + color_info = "\n".join([ + f"RGB {color[1:]} HSB {self.rgb_to_hsb(color)[0]:03.0f} {self.rgb_to_hsb(color)[1]:03.0f} {self.rgb_to_hsb(color)[2]:03.0f} 占比 {percentage:.2f}%" + for color, percentage in sorted_colors + ]) + + # draw colors image + ret_image = Image.new('RGB', size=(grid_width, grid_height * len(main_colors)), color="white") + draw = ImageDraw.Draw(ret_image) + + # Use default font with size 20 + font = ImageFont.load_default().font_variant(size=20) + + for j, (color, percentage) in enumerate(sorted_colors): + x1 = 0 + y1 = grid_height * j + draw.rectangle((x1, y1, x1 + grid_width, y1 + grid_height), fill=color, outline=color) + + # Calculate contrast color + contrast_color = self.get_contrast_color(color) + + # Add text with contrast color and HSB values + h, s, b = self.rgb_to_hsb(color) + text = f"RGB {color[1:]} HSB {h:03.0f} {s:03.0f} {b:03.0f} {percentage:.2f}%" + # 使用 font.getbbox() 来获取文本的边界框 + bbox = font.getbbox(text) + text_height = bbox[3] - bbox[1] + + # 计算文本的垂直位置，使其在色块中垂直居中 + text_x = 10 # 固定左边距为10像素 + text_y = y1 + (grid_height - text_height) // 2 + + # 绘制文本 + draw.text((text_x, text_y), text, fill=contrast_color, font=font) + + ret_images.append(pil2tensor(ret_image)) + + log(f"{self.NODE_NAME} Processed {len(ret_images)} image(s).", message_type='finish') + return (torch.cat(ret_images, dim=0), sorted_colors[0][0], sorted_colors[1][0], sorted_colors[2][0], sorted_colors[3][0], sorted_colors[4][0],) + + def ndarrays_to_colorhex(self, colors: list) -> list: + return [RGB_to_Hex((int(color[0]), int(color[1]), int(color[2]))) for color in colors] + + def interrogate_colors(self, image: torch.Tensor, num_colors: int, algorithm: str, mix_iter: int, + random_state: int) -> tuple: + from sklearn.cluster import KMeans + pixels = image.view(-1, image.shape[-1]).numpy() + kmeans = KMeans( + n_clusters=num_colors, + algorithm=algorithm, + max_iter=mix_iter, + random_state=random_state, + ).fit(pixels) + + colors = kmeans.cluster_centers_ * 255 + + # Count pixels in each cluster + labels = kmeans.labels_ + label_counts = Counter(labels) + total_pixels = len(labels) + + # Calculate percentages + color_percentages = [label_counts[i] / total_pixels * 100 for i in range(num_colors)] + + return colors, color_percentages + + def get_contrast_color(self, hex_color): + # Convert hex to RGB + rgb = tuple(int(hex_color[i:i + 2], 16) for i in (1, 3, 5)) + + # Calculate luminance + luminance = (0.299 * rgb[0] + 0.587 * rgb[1] + 0.114 * rgb[2]) / 255 + + # Choose black or white based on luminance + if luminance > 0.5: + return "#000000" # Black for light backgrounds + else: + return "#FFFFFF" # White for dark backgrounds + + def rgb_to_hsb(self, hex_color): + # Convert hex to RGB + rgb = tuple(int(hex_color[i:i + 2], 16) for i in (1, 3, 5)) + + # Convert RGB to HSB + h, s, v = colorsys.rgb_to_hsv(rgb[0] / 255, rgb[1] / 255, rgb[2] / 255) + + # Convert to degrees and percentages + h = h * 360 + s = s * 100 + b = v * 100 + + return h, s, b + + class LS_GetMainColors: def __init__(self): @@ -83,9 +225,12 @@ class LS_GetMainColors: NODE_CLASS_MAPPINGS = { - "LayerUtility: GetMainColors": LS_GetMainColors + "LayerUtility: GetMainColors": LS_GetMainColors, + "LayerUtility: GetMainColorsV2": LS_GetMainColorsV2, } NODE_DISPLAY_NAME_MAPPINGS = { - "LayerUtility: GetMainColors": "LayerUtility: Get Main Colors" + "LayerUtility: GetMainColors": "LayerUtility: Get Main Colors", + "LayerUtility: GetMainColorsV2": "LayerUtility: Get Main Colors V2", + } \ No newline at end of file diff --git a/pyproject.toml b/pyproject.toml index 1c4c4ee..1a3188b 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -1,7 +1,7 @@ [project] name = "comfyui_layerstyle" description = "A set of nodes for ComfyUI it generate image like Adobe Photoshop's Layer Style. the Drop Shadow is first completed node, and follow-up work is in progress." -version = "2.0.5" +version = "2.0.6" license = "MIT" dependencies = ["numpy", "pillow", "torch", "matplotlib", "Scipy", "scikit_image", "scikit_learn", "opencv-contrib-python", "pymatting", "timm", "colour-science", "transformers", "blend_modes", "huggingface_hub", "loguru"]